Abstract:BackgroundThe efficient and accurate diagnosis of pulmonary adenocarcinoma before surgery is of considerable significance to clinicians. Although computed tomography (CT) examinations are widely used in practice, it is still challenging and time-consuming for radiologists to distinguish between different types of subcentimeter pulmonary nodules. Although there have been many deep learning algorithms proposed, their performance largely depends on vast amounts of data, which is difficult to collect in the medical imaging area. Therefore, we propose an automatic classification system for subcentimeter pulmonary adenocarcinoma, combining a convolutional neural network (CNN) and a generative adversarial network (GAN) to optimize clinical decision-making and to provide small dataset algorithm design ideas.MethodsA total of 206 nodules with postoperative pathological labels were analyzed. Among them were 30 adenocarcinomas in situ (AISs), 119 minimally invasive adenocarcinomas (MIAs), and 57 invasive adenocarcinomas (IACs). Our system consisted of two parts, a GAN-based image synthesis, and a CNN classification. First, several popular existing GAN techniques were employed to augment the datasets, and comprehensive experiments were conducted to evaluate the quality of the GAN synthesis. Additionally, our classification system processes were based on two-dimensional (2D) nodule-centered CT patches without the need of manual labeling information.ResultsFor GAN-based image synthesis, the visual Turing test showed that even radiologists could not tell the GAN-synthesized from the raw images (accuracy: primary radiologist 56%, senior radiologist 65%). For CNN classification, our progressive growing wGAN improved the performance of CNN most effectively (area under the curve =0.83). The experiments indicated that the proposed GAN augmentation method improved the classification accuracy by 23.5% (from 37.0% to 60.5%) and 7.3% (from 53.2% to 60.5%) in comparison with training methods using raw and common augmented images respectively. The performance of this combined GAN and CNN method (accuracy: 60.5%±2.6%) was comparable to the state-of-the-art methods, and our CNN was also more lightweight.ConclusionsThe experiments revealed that GAN synthesis techniques could effectively alleviate the problem of insufficient data in medical imaging. The proposed GAN plus CNN framework can be generalized for use in building other computer-aided detection (CADx) algorithms and thus assist in diagnosis.

Automated 3D convolutional neural network architecture design using genetic algorithm for pulmonary nodule classification

Automated Lung Nodule Detection and Classification Using Deep Learning Combined with Multiple Strategies

Genetic Deep Learning for Lung Cancer Screening

Automated Detection and Classification for Early Stage Lung Cancer on CT Images Using Deep Learning

Lung Nodule Classification of CT Images Based on the Deep Learning Algorithms

An improved CNN-based architecture for automatic lung nodule classification

Lung Nodule Diagnosis on 3D Computed Tomography Images Using Deep Convolutional Neural Networks

Deep 3D Convolutional Neural Network for Automated Lung Cancer Diagnosis

Medical AI for Early Detection of Lung Cancer: A Survey

Automated Pulmonary Nodule Detection in CT Images Using Deep Convolutional Neural Networks.

Detection and Classification of Pulmonary Nodules Using Convolutional Neural Networks: A Survey

A Novel Self-Adaptive Convolutional Neural Network Model Using Spatial Pyramid Pooling for 3D Lung Nodule Computer-Aided Diagnosis

Lung Nodule Classification on CT Images Using Deep Convolutional Neural Network Based on Geometric Feature Extraction

Lung Cancer Detection and Severity Analysis with a 3D Deep Learning CNN Model Using CT-DICOM Clinical Dataset

An Improved Convolution Neural Network and Modified Regularized K-Means-Based Automatic Lung Nodule Detection and Classification

Lung Nodule Classification of CT Images Using Channel and Spatial Attention CNN with Bayesian Optimization

Toward an Expert Level of Lung Cancer Detection and Classification Using a Deep Convolutional Neural Network.

Research of Lung Cancer-Assisted Diagnosis Algorithm Based on Multi-scale Convolution Kernel Network

Combination of Generative Adversarial Network and Convolutional Neural Network for Automatic Subcentimeter Pulmonary Adenocarcinoma Classification.

Fast-staged CNN Model for Accurate pulmonary diseases and Lung cancer detection

An Efficient Model for Lungs Nodule Classification Using Supervised Learning Technique